Conduit heats up

Status
Not open for further replies.

bikeindy

Senior Member
Location
Indianapolis IN
Was at a commercial building and the owner asked me why the conduit from his switch boxes were getting warm to almost hot. I took a quick look at it and there were 2 sets of 3 way switches one feeding overhead lights and one feeding fans and a 2 pole switch feeding two more sets of overhead lights. total of 4 circuits run in two 3/4" EMT conduit. power is 120/208 3 phase and there were 2 grounded conductors. the quick look i took it appears that the circuits are phased properly to the grounded conductors and the pipe is not at all overloaded with wire the heat stops about 15 feet from the switch box. each 20 amp circuit was feeding 4 400watt high bay MH lights. I'd like to solve this for him but was a little confused as to what is causing the heat.
 
O

oliver100

Guest
In a magnetic based material conduit, the sum of all participating current phase vectors (neutral included) must be ZERO, i.e. all current supplied to the load through the conduit must close the loop to the source through the same conduit.
 

charlie tuna

Senior Member
Location
Florida
there would be two reasons for this raceway to be warm or hot:

A.) the raceway could be returning some sort of fault or neutral
current back to the power source.

B.) load and conductor fill

from the loads you listed i believe it is B.

i remember a service call years ago where a pvc manufacturer called us about three hot raceways in his plant. we found the problem to be caused by someone installing separate feeders to ranges and used the raceway as a neutral return since he did not pull a third wire to each range. when conductors are crowded (but still legal) in a raceway and running near 80 per cent of capacity, the total amount of heat generated can be alarming.......
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
Re: Conduit heats up

bikeindy said:
. . . the heat stops about 15 feet from the switch box.
That's the symptom that I find most confusing. If the conduit had been hot for its entire length, then my "prime suspects" would be, (1) Improper grouping of wires in the two conduits, such as all Phase A in one and all Phase B in the other, (2) Too many wires in a conduit, and (3) Too much current in one or more of the wires. But your investigation seems to have vindicated those suspects.

One question: is it only one conduit that is getting hot, or are both?

The only other possibility I can think of is that there might be a short circuit inside the conduit, and that the conduit is carrying current from the fault point back to the source. But I should think that that failure would make itself known by other means, such as giving anyone who touches it a shocking feeling.
 

tshea

Senior Member
Location
Wisconsin
I took a quick look at it and there were 2 sets of 3 way switches one feeding overhead lights and one feeding fans and a 2 pole switch feeding two more sets of overhead lights.

Is the switch box hotter than the conduit? If so I'll guess that you have 15A switches and these should probably be 20A switches (as in heavy duty/commercial grade) and not the cheap ones.


the heat stops about 15 feet from the switch box.

The conduit is acting as a heat sink and the furthur away from the heat source the less warm it is.
 

bikeindy

Senior Member
Location
Indianapolis IN
oliver100 said:
In a magnetic based material conduit, the sum of all participating current phase vectors (neutral included) must be ZERO, i.e. all current supplied to the load through the conduit must close the loop to the source through the same conduit.

I think we are on the right track here. Conduit fill is in no way a problem as there are only 6 conductors in each conduit. . There are no shorts and the "HOT' is not voltage on the conduit it is temperature. The switches are high quality 20 Amp switches yes the box may be a little warmer than the conduit but you would have to use a temp gauge to notice it I think. there are no "neutral" conductors from the J-box 5 feet from the switches to the switch box. they just feed through at the J-box. the conduit cools starting at the J-box in both directions. so at that point I think the conduit is acting as a heat sink. What do you think Oliver100
 

charlie b

Moderator
Staff member
Location
Lockport, IL
Occupation
Retired Electrical Engineer
bikeindy said:
. . . there are no "neutral" conductors from the J-box 5 feet from the switches to the switch box. they just feed through at the J-box.
Here's what I think the current path should be between the two boxes: Current leaves the j-box on one wire, goes through the switch, and returns to the j-box on another wire. You should see the same current in the two wires, except that the two currents are in opposite directions. Their magnetic fields should essentially cancel each other out, and should not cause any heating of the conduit.

But do you have some type of wiring arrangement that is different from what I describe above? If so, that might well be the cause of the heating problem. Can you describe the wiring between the j-box and the switch box (i.e., number and character of wires)?
 

bikeindy

Senior Member
Location
Indianapolis IN
from the J- Box there are two 3/4" conduits to the Switch Box, two 3 way switches which have the travelers and switch legs are in one of the conduits and are on separate 20 Amp circuits A and B phase. in conduit # 2 there are two 20 Amp circuits power feed and switch legs to a two pole 20 amp switch Phase C and A. Conduit 1 has 6 #12 thhn wire and conduit 2 has 4 #12 thhn wires. There are 2 "neutrals in the J-box feeding through and are properly phased and one is a #10 THHN for the legs that are phase A B and C.
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
charlie b said:
bikeindy said:
. . . there are no "neutral" conductors from the J-box 5 feet from the switches to the switch box. they just feed through at the J-box.
Here's what I think the current path should be between the two boxes: Current leaves the j-box on one wire, goes through the switch, and returns to the j-box on another wire.
One thing we need to keep in mind. Unless the neutrals follows the hots down to the switches and back, which they don't here, for each wire that travels down to a switch, the wire returning that current must be inside the same conduit, otherwise we get eddies.

In other words, you can't have the hots feeding down one conduit, hit the switches, and then travel back up the other conduit. For each 3-way, the common and two travelers must be in one conduit. For the 2-pole switch, all four wires must be in one conduit.

If there are two 3-ways and a 2-pole, that should be a total of ten wires. We need a recount.
 

LarryFine

Master Electrician Electric Contractor Richmond VA
Location
Henrico County, VA
Occupation
Electrical Contractor
Okay, I guess I type slowly. The wire count seems correct. One other thought: 1600 watts per circuit is a full load, and I would expect some heat with 4 current-carrying conductors per pipe.

It may be time to bring in an IR camera.
 

tonyi

Senior Member
The pipe should contain most of the EM field from the conductors inside right?

One easy test should be to just clamp a probe on the pipe and see if its carrying any current. If not, look elsewhere.
 
O

oliver100

Guest
What do you think Oliver100

The heat, that stops in the middle of the conduit is the most confusing thing. Is it heated by nearby wiring, the sun or anything else ??? :lol:

It is possible that the current balance is disrupted somewhere in the lamps or fan the wiring. Something like a load gets 208 volts from wires coming from two different conduits.

If you solve it, let us know the mistery :? :? :?
 

bikeindy

Senior Member
Location
Indianapolis IN
There is nothing but the current that can be causing this I think I just don't get it. the fixtures are twenty feet in the air about thirty-five feet of pipe from the switch to the J- box for the first set of fixtures. no direct sunlight. I think I will run a neutral down each conduit from the j_box and see what happens. I haven't climbed up to take a look the lamps are all wired 120V and every thing works well. and the pipe is not hot to the touch it is pretty warm though. been going on for two years he said and the wires show no sign of being over heated. there is no other wiring near by. I am hoping to get this guy fixed up but I told him I would need time and a lift and it could cost him.
 

tshea

Senior Member
Location
Wisconsin
Can you work the circuit before the lights are on and the pipe gets hot?
If so, remove the switches from teh box and turn on the circuit.
The conduit will be cold. within 15 minutes I bet the switches are getting warm. My bet is the switches are dissipating heat into the conduit.

I would then replace the switches if my guess is right.
 

izak

Senior Member
Location
Springfield, MO
ive run into conduit heating problems when the 'neutral' wasnt in the pipe with the hots or travelers

seems like the 'neutrals' for those 3 ways ought to be in the pipe with those travelers.

any takers?
 

mhulbert

Senior Member
Location
Chico, CA
It's not about neutrals...you just have to have a "balance" of current coming and going through the raceway. In other words, if you have 12A going from left to right, you better have 12A going right to left. If the pipe feed, say, a light, we have a hot and a neutral. The current in the hot goes one way, and the exact same amount of current goes the other way through the neutral. The magnetic field is thus canceled out.

But it doesn't matter if it's a neutral or not...in the case of the opening post, we have a typical "dead end" 3-way. we have current going down, to the switch in the supply lead, and we have current going back up in either of the travelers(depending on the position of the switch). Adding a neutral (and dead ending it in the switch box) won't do a thing, there is no current flow in the neutral. Now, what if the supply wire came down and fed a receptacle as well? We'd have to have a neutral to cancel out the excess current drawn by a receptacle (and make the receptacle work). For example, say the light that the 3-way powered needed 5A, and the receptacle had a load that needed 3A. The supply lead will carry 8A down the pipe, the traveller will carry 5A up the pipe, and the neutral will carry 3A. Everything balances out: 8A down, 8A up. If you don't have equal amounts of current going in each direction, you will have induction heating.

Maybe this will clear up the confusion with adding a useless nuetral wire.

mike
 

hardworkingstiff

Senior Member
Location
Wilmington, NC
From the OP, it appears that the neutrals may be useless, but my understanding is they pass through. We don't know how they are connected. I suspect someone has them connected in a way that the neutral does have a load on it that is not countered by the current of the ungrounded conductor. It is difficult to know what the OP is really saying about the neutral.
 

bikeindy

Senior Member
Location
Indianapolis IN
tried to be plain as I could. The neutrals pass through the J-Box and never enter the switch Box. So I was by there today as a gesture of concern for this guy and his building. I am sure this will be great for me in terms of work since he is now telling all his customers to call me. Its a hot tub place. OK for you who want to blame the switches sorry I wire nutted and eliminated them for a time to see what would happen. ran a neutral down and back up to the J- box in each conductorconnected them in the box as so they were in a series loop so voltage had to pass through them. and. :x :evil: :?: :!: :roll: :cry: :lol: STILL WARM CONDUIT!!! I am 99% sure the wiring is right and So we will think some more and go back later. the guy said If I fix the problem he will pay for all my time and a lift if nessasary he said I could forget it if I wanted. I started looking at it because I was there as a customer and thought I could help him out now I have a great contact for work. I am more interested in figuring it out at this point just to know what the deal is with it. remember this has been this way for 2 years and the wire looks new. very strange. turn the lights out and the heat leaves in about ten minutes.
 
Status
Not open for further replies.
Top